Test Suite Quality for Model Transformation Chains

Abstract

For testing model transformations or model transformation chains, a software engineer usually designs a test suite consisting of test cases where each test case consists of one or several models. In order to ensure a high quality of such a test suite, coverage achieved by the test cases with regards to the system under test must be systematically measured. Using coverage analysis and the resulting coverage information, missing test cases and redundant test cases can be identified and thereby the quality of the test suite can be improved. As test cases consist of models, a coverage analysis approach must measure how complete models cover the domains of the transformations in the chain and to what degree of completeness transformations are covered when executing the test suite. In this paper, we present a coverage analysis approach for measuring test suite quality for model transformation chains. Our approach combines different coverage criteria and yields detailed coverage information that can be used to identify missing and redundant test cases.